The invention relates the crystal structure of IMPDH isolated from bacteria. The structure is different from the structure of mammalian or fungal IMPDH, allowing design of selective inhibitors of bacterial IMPDH.
Inosine monophosphate dehydrogenase (IMPDH; Enzyme Commission (EC) 1.1.1.205) is a rate-limiting enzyme in the synthesis of guanine ribonucleotides. IMPDH has an essential role in providing critical precursors for DNA and RNA biosynthesis and in signal transduction pathways that mediate cell differentiation (Collart et at., 1990; Kiguchi et al., 1990). Because of its central role in purine metabolism, IMPDH is an attractive therapeutic target. Several recent reviews have outlined the utility of mammalian IMPDH inhibitors as anticancer (Pankiewicz, 1997) or antiviral (Andrei et al, 1993) agents or as immunosuppressive drugs (Halloran, 1996) (see Table 1).
Although there are no reports of selective inhibitors of bacterial IMPDH enzymes, such compounds could have potential application as specific antimicrobial agents.
The active form of IMPDH enzymes (50-55 kDa) is a homotetramer with four active sites per tetramer. A cysteine residue in the active site forms a covalent intermediate with IMP (Wang et al., 1996). A consensus sequence of thirteen amino acid residues that includes cysteine in this active site has been proposed as a signature motif (i.e., an amino acid sequence that can be used as a fingerprint or specific identifier for this class of enzymes) for the IMPDH and guanosine monophosphate (GMP) reductase enzymes (Bairoch, 1995). This IMPDH consensus region is highly conserved in both bacteria and eukaryotes, with 90% and 85% of the respective residues being identical within each kingdom. However, only 40% of these residues remain identical when compared between the two kingdoms. This limited conservation suggests that bacterial and eukaryotic IMPDH enzymes may have distinct characteristics; a suggestion supported by their kinetic differences and differential sensitivity to inhibitors. Enzymes from mammalian sources show distinctly lower values for the Km for nicotinamide adenine dinucleotide (NAD) than do those enzymes from bacteria. In addition, mammalian IMPDH enzymes are 10-100 times more sensitive to inhibition by mycophenolic acid (MPA) than are bacterial IMPDH enzymes. Sequence analysis of all known IMPDH enzymes supports the distinction between bacterial and eukaryotic enzymes. A deep branching of the bacterial and eukaryotic forms of IMPDH is observed upon phylogenetic analysis of the relationships among the various IMPDH genes (Collart et al, 1996 a and b). This phylogenetic analysis indicates a general functional conservation of amino acid and suggests a unique amino acid sequence signature for these kingdoms.
The elucidation of a kingdom-specific signature for IMPDH enzymes is an important element in the development of specific inhibitors. The two partial structures of IMPDH from Chinese hamster (Sintchak et al., 1996) (85% structure complete with bound transition state analogue and mycophenolic acid, MPA) and Tritrichomonas foetus (Whitby et al., 1997) (68% structure complete with bound xanthosine monophosphate [XMP]) have been reported with only the coordinates of the latter available in the Protein Data Bank (PDB). These structures furnished the initial information about the structure and reaction mechanism of eukaryotic IMPDH enzymes. Inhibitors of IMPDH in bacteria are needed to treat infections, in particular, to overcome the barrier of antibiotic resistance.
The invention relates for the first time a crystal structure of a bacterial IMPDH. This invention relates that bacterial and mammalian IMPDH enzymes provide the same catalytic function, but have a set of unique structural and biochemical characteristics. An embodiment is a crystal structure of IMPDH isolated from Streptococcus pyogenes. S. pyogenes IMP dehydrogenase represents the class of bacterial IMPDH enzymes that show distinct functional differences when compared to mammalian IMPDH enzymes. The bacterial enzymes bind NAD poorly (Zhou et al., 1997; Kerr et al., 1997) (KM greater than 1 mM) and are inhibited by MPA only at very high concentrations (Ki greater than 0.5 mM). Elucidation of the structural basis of these distinct characteristics is useful to aid in design of specific IMPDH inhibitors that will inhibit the infectious agent without harming the host""s IMPDH.
The coding sequence of bacterial IMPDH specifies a protein of 493 amino acids that contain only a single cysteine residue at the active site (Ashbaugh et al., 1995). IMPDH from S. pyogenes is a representative bacterial enzyme because the organism is pathogenic, and therefore a good model for the investigation of enzyme inhibitors. Streptococci are the most common cause of worldwide pneumonia and a leading cause of pediatric infections. The structure of the S. pyogenes bacterial IMPDH provides the basis for elucidation of the structural characteristics that distinguish bacterial from eukaryotic IMPDH enzymes. Knowledge of these characteristics permits an understanding of why these enzymes exhibit functionally distinct behavior and therefore provides a foundation for the design of specific inhibitors of IMPDH that have clinical value.
In addition to inhibiting pathogens, the immunosuppressive use of IMPDH inhibitors is applicable to treat chronic inflammatory diseases such as arthritis, diabetes, or systemic lupus erythromotosis. Use of the IMPDH structure from S. pyogenes will facilitate identification of other pathogens that will be inhibited by drugs that inhibit S. pyogenes. 
Definitions and Abbreviations
A xe2x80x9cbinding pocketxe2x80x9d is a space in a molecule in which an inhibitor of the molecule is bound.
The following abbreviations are used throughout the application: